Thuc Tran

🚀 We’re partnering with top software engineers in New York’s early-stage startup scene. Talent is moving fast, but our clients are listening and adapting as engineers face real challenges: Interview fatigue is real. Engineers are burned out from marathon processes, long take-homes, and ghosting feedback. “We don’t get to see the product”. Companies aren’t demoing or explaining the product early on, and it’s a big turn-off. The best candidates want to believe in what they’re building. Show, don’t just tell. Tech rounds are moving up. More companies are front-loading the technical interviews to avoid time wasted later, and it’s working for both sides. Remote is shrinking. We’re seeing a definite tilt back toward onsite or hybrid. Not everyone loves it, but the shift is happening. It’s less about a coding language, more about product. Hiring managers are becoming less rigid about specific stacks, and more focused on alignment with product vision, user impact, and client collaboration. 📩 Reach out for more insights: george.wilson@stottandmay.com #Hiring #SoftwareEngineering #TechRecruiting

22

1 Comment

Why "On-Device" is the most ethical AI strategy. We talk a lot about the size of models. Parameters. Trillions of tokens. The race for AGI. But in my academic life at EPFL and my work in Speech, I’ve always been obsessed with the opposite end of the spectrum: Accessibility. When we published our research in Nature Communications on AI for social projects, one conclusion was clear: If AI requires a massive data center and a high-speed fiber connection to run, it excludes half the world. Cloud models are undeniably impressive and they are pushing the boundaries of what is possible. But they shouldn't be the only way we deploy intelligence. I'm seeing a resurgence of interest in on-device solutions, not to replace the cloud, but to handle the sensitive, immediate, and personal tasks where the edge shines: - Privacy: Processing data right where it’s created. This is crucial for multimodal data (voice, video) that is highly identifiable and shouldn't leave the user's hand. - Latency: Real-time processing that feels like a conversation, not a transaction. - Energy Efficiency: Running a small, optimized model locally is often far more sustainable than round-tripping every query to a power-hungry data center. - Reliability: True accessibility means working offline. AI should be helpful even when the signal drops or the fiber hasn't been laid yet. Better locally-run alternatives make these powerful tools accessible to everyone, even for problems and data that are highly sensitive. I’m planning to dive deeper into this myself by experimenting with some of the latest open-weights models specifically for local deployment. If you’ve been testing small language models (SLMs) or edge-optimized architectures recently, I’d love to hear about your experience. What frameworks or models are you finding most promising for mobile/edge/small server hardware right now? #AIforGood #OnDeviceML #NLP #EthicalAI #TechForGood #EdgeAI #SustainableTech #OpenWeights

7

6 Comments

The $100,000 H-1B visa fee is now in effect. At the very moment the U.S. is in a race to dominate AI, access to global talent is being cut off. NVIDIA chips are already restricted. Compute is scarce. Talent in AI engineering and research is the other critical bottleneck. Now that pipeline has been narrowed too. For hiring managers, this is not just an immigration issue, it is a competitiveness issue: • Smaller companies will struggle to access the skills needed to build in AI • Global graduates may look to Canada, Europe, or Asia instead of the U.S. • Talent scarcity at home will push salaries even higher and widen the skills gap Innovation does not stop because of policy. It relocates. By making it harder for top global minds to work in the U.S., this decision risks slowing American leadership in AI and handing opportunity to other markets. What do you think… does this protect jobs, or does it put the U.S. at risk of falling behind in the most important technology race of our time?

26

16 Comments

Etched just raised ~$500M in a major funding round (valued at ~$5B) to accelerate its custom AI silicon — a solid alternative to Nvidia-centric stacks. https://lnkd.in/dttxqJPJ

6

Bad day to be a Waymo, and even worse to be stuck inside or behind one. A widespread blackout stalled robotaxis across the city Saturday night, turning “the future of mobility” into traffic chaos. Service has since been suspended. This isn’t really about Waymo. It’s about what happens when energy-intensive, always-on technology collides with aging infrastructure that already struggles under peak demand. We keep talking about an AI and automation revolution as if it’s mostly software. It isn’t. These systems depend on stable power, resilient grids, and redundancy that many cities simply do not have yet. When the electricity goes out, the future literally stops in the middle of the street. What makes this more frustrating is that the problem is not new. A major federal infrastructure bill was passed years ago with explicit goals around grid modernization and resilience. The money exists. The need is obvious. But implementation has been slow, fragmented, and largely invisible, while the tech rollout races ahead. This gap matters. We’re layering autonomous vehicles, data centers, AI workloads, and electrification onto foundations that were not designed for this level of continuous demand. The question isn’t whether this technology is impressive. It is. The question is whether we’re building the boring, unglamorous infrastructure fast enough to support it. Right now, the answer looks a lot like no. https://lnkd.in/eTN4xEK3

2

9 Comments

Every AI model you've ever used ran on a chip. That chip took 2-3 years and hundreds of millions of dollars to design. Partcl is making that process 7,000x faster. Not a typo. They're a 3-person team out of YC, backed by top VC, building GPU-accelerated EDA tools that compress what used to take months into hours — using custom diffusion models and reinforcement learning trained on the hardest physical layout problems in chip design. In the last 18 months their first engineer: → Built a physics simulator that matched state-of-the-art legacy tools in 6 weeks (experts said it would take 10 years) → Took a process running in 5 minutes down to 20 seconds → Pushed their speedup from 700x to 7,000x They're hiring Founding Engineer- This is the role for someone who: — Has trained custom ML models from scratch (robotics, physics-informed, biophysics — not fine-tuning LLaMA) — Thinks in GPU memory layouts and kernel optimization, not just PyTorch APIs — Wants to be in the room when the VP of Engineering at NVIDIA gives feedback on something they personally built — Would rather rebuild something from scratch than maintain it Stack: PyTorch, Rust, Triton, CUDA, C++ Location: San Francisco, in-person The market is $15B. And they have paying customers. If you write CUDA kernels for fun, have trained physics-informed models, or built something from zero that no one thought was possible — I'd love to talk. DM me or drop a link to your GitHub.

5

Alibaba-backed startup Moonshot : open-sourced Kimi K2.5, a new 1T-parameter model that rivals GPT-5.2 and Claude Opus 4.5 across coding, vision, and agentic benchmarks. The release includes Kimi Code, a terminal-based coding agent Founder, Zhilin Yang first time speaking on camera and shared Kimi K2.5 Why this matters: • Native multimodal capabilities — K2.5 understands and generates across text, images, and video, enabling truly visual workflows. • Visual coding excellence — turn screenshots or videos into frontend code with precision, bridging design and development. • Agent Swarm intelligence — orchestrate up to 100 autonomous agents working in parallel to tackle complex tasks faster and at scale. • Real-world productivity — expert-level handling of docs, spreadsheets, slides, and long-form outputs, accelerating everyday professional work. Kimi K2.5 represents a meaningful leap toward AI systems that do more than generate text — they see, reason, code, and collaborate like a team of specialists. Explore more at Kimi.com and see how open-source AI is redefining what’s possible. https://lnkd.in/gS2zjVUN #ArtificialIntelligence #OpenSourceAI #MachineLearning #MultimodalAI #Innovation

3